EricRollei-AAA_Metadata_System/documentation/ReadMe Docs/eric_metadata_save_image_node-documentation.md

Save Image with Metadata Node

Overview

The "Save Image with Metadata" node extends ComfyUI's standard image saving capabilities with comprehensive metadata handling. It captures AI generation parameters, workflow information, and user-provided descriptive metadata, then embeds this information directly into saved images.

This node is designed to create well-documented images that preserve their creation history for archiving, organization, and future reference.

Installation & Dependencies

Requirements:

• ComfyUI
• Metadata_system package
• Python Imaging Library (PIL/Pillow)
• PyExiv2 (for metadata handling)

Installation:

1. Ensure the Metadata_system package is installed in your ComfyUI environment
2. Place the node file in your ComfyUI custom_nodes directory
3. Restart ComfyUI

How It Works

Comprehensive Metadata Structure:

• The basic metadata section includes fields for title, description, copyright, creator, and project
• Keywords are properly parsed from comma-separated input
• Rights metadata follows standard XMP conventions

AI Generation Information:

• Automatically captures generation parameters (model, prompts, sampler, seed, etc.)
• Preserves workflow information, identifying active nodes that contributed to the output
• Includes summary information about the workflow complexity

Advanced File Handling:

• Supports custom output directories
• Automatically formats filenames with datetime and project information when enabled
• Maintains compatibility with ComfyUI's folder structure

User Flexibility:

• All metadata fields are optional - you can fill in as much or as little as you want
• Boolean toggles for including datetime/project in filenames
• Supports multiple metadata storage formats (embedded, XMP sidecar, text files)

Input Parameters

Required Inputs:

• images: The image tensor to save
• filename_prefix: Base prefix for the output filename

Optional Inputs:

Basic Metadata:

• title: Image title
• description: Detailed image description
• keywords: Comma-separated keywords for categorization
• copyright: Rights information
• project: Project name (useful for grouping related images)
• creator: Author/creator name

Generation Parameters:

• model_name: Name of the model used
• positive_prompt: Positive generation prompt
• negative_prompt: Negative generation prompt
• seed: Generation seed
• steps: Number of generation steps
• cfg_scale: CFG scale value
• sampler: Sampler name
• scheduler: Scheduler name

File Handling:

• custom_output_directory: Path to custom save location
• include_datetime: Add date/time to filenames (boolean)
• include_project: Add project name to filenames (boolean)

Workflow Options:

• workflow_data: Explicit workflow JSON data
• enable_metadata: Toggle metadata writing (boolean)
• metadata_targets: Comma-separated list of targets (embedded,xmp,txt)
• custom_metadata: JSON string with custom metadata
• auto_capture_workflow: Attempt to capture workflow automatically (boolean)
• workflow_id: Specific workflow ID to use

Technical Details

Metadata Standards:

• XMP: Adobe's Extensible Metadata Platform for standardized metadata
• IPTC: International Press Telecommunications Council standard
• EXIF: Exchangeable Image File Format for camera data

Storage Locations:

• Embedded: Metadata stored directly inside image files
• XMP Sidecar: Companion XML files with .xmp extension
• Text Files: Human-readable text files with .txt extension
• Database: Optional storage in SQLite database (if configured)

Workflow Capture Methods:

1. Direct Input: From workflow_data parameter (highest priority)
2. ComfyUI Context: Automatically captured from current execution (when enabled)
3. Active Node Analysis: Identifies which nodes contributed to the final output

Troubleshooting

Common Issues:

Metadata Not Appearing:

• Ensure enable_metadata is set to True
• Check that appropriate targets are included in metadata_targets
• Verify you have writing permissions for the output location

Workflow Data Missing:

• Enable auto_capture_workflow
• Check ComfyUI version compatibility
• Try providing explicit workflow data via connection

Invalid Output Directory:

• Ensure custom directory path exists or can be created
• Check permissions for the specified location
• Use absolute paths to avoid ambiguity

Advanced Usage

Custom Metadata Format:

You can provide additional custom metadata via JSON. Example format:

{
  "basic": {
    "rating": 5
  },
  "analysis": {
    "technical": {
      "quality_score": 0.95
    }
  }
}

Integration with Other Nodes:

• Connect text nodes to metadata fields
• Use output from analysis nodes as input to custom_metadata
• Chain with other metadata-aware nodes for complete workflows

Batch Processing:

When processing batched images, the node will add sequential numbering to filenames automatically and apply the same metadata to each image in the batch.